Cigarette smoking in adolescents is a major public health problem, which not only relates to the younger population, but also for the long-term physical and mental health of adults. Studies suggest that most adults could be prevented from becoming tobacco users if they could be kept tobacco-free during adolescence. However, not all youth who begin smoking develop nicotine dependence (ND). In order for the prevention efforts for smoking to be most effective, it is important to elucidate the underlying mechanisms that drive the progression to ND, as well as to gain a better understanding of the individual differences in vulnerability. Data from diverse lines of research suggest a close connection between depression and ND. However, the mechanisms underlying the linkages between these two disorders are poorly understood, thereby limiting the development of effective treatments. Studies in both humans and animals suggest that the stress-response system may be involved in both ND and depressive disorder. Using a longitudinal design, this investigation seeks to examine the association between hypothalamic-pituitary-adrenal (HPA) dysregulation and risk for smoking initiation and progression to ND in four groups of adolescents: (1) depressed adolescents with no previous or current tobacco use; (2) depressed adolescents with current tobacco use; (3) normal adolescents with no previous or current tobacco use; and (4) normal adolescents with current tobacco use. Pilot data from our laboratory have shown that elevated cortisol near bedtime increases the risk for non-nicotine substance-related disorders and for recurrent depressive episodes in depressed adolescents. Other data suggest that antidepressant (AD) agents are helpful in reducing nicotine/other substance use in addition to improving depressive symptoms, which might be due to down-regulation of the HPA axis by the AD drugs. If our preliminary results, showing a relationship between high cortisol and risk for substance use disorders, are extended to ND, the findings potentially might be helpful in identifying the subgroup of youth with these disorders who might benefit most from AD or anti-glucocorticoid treatment strategies. This would have important clinical implications because, as a group, youngsters tend to show more variability in AD response compared with adult depressed patients. More focused and effective interventions early in the course of the disorder might be helpful in reducing the long-term morbidity and mortality associated with ND. Even if only a small subgroup of youngsters at risk for ND can be identified by such means, because of the many direct as well as indirect negative consequences associated with ND, this will be of great public health importance.